Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Winstar Comm. (WCII) -- Ignore unavailable to you. Want to Upgrade?

To: Bernard Levy who wrote (4765)	3/27/1998 6:50:00 PM
From: James Fink	Read Replies (2) \| Respond to of 12468

Man, I wish I was as smart as Bernard Levy!

To: Bernard Levy who wrote (4765)	3/27/1998 9:10:00 PM
From: SteveG	Read Replies (1) \| Respond to of 12468

Hi Bernard- Thanks again for another textbook response. More questions: <..a fixed amount of bandwidth to each point to point link, independently of the amount of traffic on this link. Point to multipoint transmission has the ability to allocate bandwidth based on the actual demand, so that it increases capacity...> Is the P-MP bandwidth allocation then more like ethernet, or does it have some QoS capabilities like VBR and CBR? And does P-P in general have the capability to accomodate bursts, or is it just that WCII may have chosen to limit the P-P bandwidth? And I'm pretty elementary in my understanding of the P-P/P-MP distinctions, but I have seen reference to P-MP being "digital" relative to P-P. Is P-P analog? Is the main benefit of P-MP that a receiver can retransmit the signal to another non-LOS receiver? Rouhana made the point that P-MP will increase their potential plant by 15-20x, which I uderstood was a function of this non-LOS P-MP capability, presumably sectorized. <..one for each of the orthogonally phase-shifted modulating waveforms..> I have understood that using "polarization" techniques (which I'm not sure if this is simple 2D phase shifting or an orthogonal phase shift or something completely different), FDM and WDM systems can double their bandwidth. I know Lucent applies some form of this in their DWDM, and that it is theoretically possible for wireless. Just don't know any specifics. <..discrete multitone (DMT) and discrete wavelet multitone schemes for ADSL/VDSL are extremely interesting since they come very close to achieving the theoretical capacity of a communication channel...> My understanding from some DSL engineers is as you suggest, ADSL (both DMT and CAP) pretty closely approach the Shannon channel capacity. With hardware (both ASICs and DSP) implementation soon available for ADSL, is it a short or long wait for adaptation to wireless. And do you see this as a likely R&D direction? And who would be working on stuff like this? MOT? ROK? PCMS? <..Wireless ATM is one implementation of bandwidth management. However, I am not sure that any standard has yet been put in place..> Did you see that the radios that LU is fronting to ARTT will use (YURI based, LU OEM'ed) ATM switching? Thanks again and have a great one Bernard! Steve

To: Bernard Levy who wrote (4765)	3/30/1998 9:41:00 AM
From: gauguin	Read Replies (1) \| Respond to of 12468

From what I remember, regarding how WCII deploys its point to point bandwidth...A single antenna has a minimum of 4 T-1 capacity, so even if they only sell one customer in a building needing one T1, the dish can handle more, which is why it is cost effective (given roof rights) to continue selling in that building. Bandwidth can be increased remotely thru a software download to the antenna. They now also have the ability (I believe) to sell Ds1 and Ds3 for higher bandwidth-demanding customers. Given that 38GHz only requires 2 degrees of separation between links, they can put multiples of 4T1's, DS1's (and maybe even DS3's) on the same mini-tower on a building.

To: Bernard Levy who wrote (4765)	4/1/1998 12:17:00 AM
From: SteveG	Read Replies (3) \| Respond to of 12468

<... orthogonal multicarrier modulation, which is the technique employed by the discrete multitone (DMT) and discrete wavelet multitone schemes for ADSL/VDSL are extremely interesting since they come very close to achieving the theoretical capacity of a communication channel....> In figuring Shannon channel capacity, since each frequency has a different signal and noise power spectrum, [with capacity C = log_2 ( 1 + S(f)/N(f) )bits/sec/Hz] are there theoretical SNR values for 38GHz, including "real world" approximations, to figure what additional channel capacity MIGHT be squeezed out with more complex modulation. And if we restrict modulation to 100MHz channels, do you still envision the need for GaAs, and would this work if greater bandwidth is desired by simply adding channels? Thanks again.